赞
踩
自从我们有了新的vSphere版本以来,已经有一段时间了,但是等待已经结束– vSphere 6.7已经正式发布!我一直在参与6.7 beta版,因此能够公开谈论该版本终于很高兴。我想发表一篇文章,概述此版本中已更新或新增的一些新功能。因此,让我们开始吧!
让我们先谈谈您可以升级和不能升级的内容。新版本仅支持从vSphere 6.0或6.5的升级和迁移。如果您的环境运行的是5.5版,则升级到6.7需要额外的步骤。您将必须至少迁移到6.0版,然后再迁移到6.7版。如果您具有运行6.0或6.5的vCenter Server管理ESXi主机5.5的混合环境,则必须在升级vCenter Server之前将主机至少升级到版本6.0。VMware已经为此花了很多时间和警告,很多人看到了它的到来,但是现在它是正式的,尽快脱离5.5!
当前支持的版本6.7迁移路径:
VMware再次更新了一些最大值。大多数常用的CPU和内存内容保持不变,但下面列出了其他更改。容错VM进行了另一次升级,现在可以升级到8个vCPU,这很好。
虚拟机最大值 | 6.5 | 6.7 |
永久内存–每个VM的NVDIMM控制器 | 不适用 | 1个 |
永久内存–每个虚拟机的非易失性内存 | 不适用 | 1024GB |
存储虚拟适配器和设备–每个虚拟SCSI适配器的虚拟SCSI目标 | 15 | 64 |
存储虚拟适配器和设备–每个虚拟机的虚拟SCSI目标 | 60 | 256 |
虚拟设备联网–每个虚拟机的虚拟RDMA适配器 | 不适用 | 1个 |
ESXi主机最大值 | 6.5 | 6.7 |
最大容错能力–每个虚拟机的虚拟CPU | 4 | 8 |
最高容错能力–每个FT VM的RAM | 64GB | 128GB |
主机CPU最大值–每个主机的逻辑CPU | 576 | 768 |
ESXi主机持久性内存最大值–每个主机的最大非易失性内存 | 不适用 | 1TB |
ESXi主机内存最大值–每个主机的最大RAM | 12TB | 16TB |
光纤通道–服务器上的总路径数 | 2048 | 4096 |
通用VMFS –每个主机的卷 | 512 | 1024 |
iSCSI物理–每个服务器的LUN | 512 | 1024 |
iSCSI物理–服务器上的总路径数 | 2048 | 4096 |
光纤通道–每个主机的LUN | 512 | 1024 |
虚拟卷–每个主机的PE数 | 256 | 512 |
对我来说,这是我在6.7中最喜欢的更新。我经常被问到他们何时要杀死基于Flash的客户端,而随着去年宣布弃用基于Flash的客户端,答案很快就会得到答案。HTML 5客户端现在已完成95%的功能。在过去的几个月中,他们确实为客户做了大量工作,并为之努力。他们在这里所做的工作给我留下了深刻的印象,这对vCenter团队来说真是太好了!Platform Services Controller功能现在也已成为vSphere Client的一部分。现在,它位于“管理”菜单下,“ PSC”选项位于两个选项卡之间。证书管理有其自己的标签,所有其他管理均在配置标签下。
更新的vSphere Client版本中的一些较新的工作流程包括:
请注意,并非所有上述工作流程都已100%完成,但是VMware每次更新都在不断更新工作流程。这只是时间问题。
6.5中内置了内置备份工具,但是要能够轻松自动备份设备是一个挑战。如下所示,您现在可以设置基本的备份计划,以备份设备配置以供以后还原。您还可以设置要保留的备份数量。我仍然建议您使用Altaro之类的第三方备份工具来备份vCenter Linux Appliance,因为您永远无法拥有足够的备份!
它已经回来了!vCenter 6.7的一项新功能是能够将vCenter Server重新指向驻留在完全不同的vSphere SSO域中的另一个Platform Services Controller节点 。对于域合并,此功能非常庞大,如果需要,可以使用cmss-util工具进行域拆分。
该功能在vSphere 5.5中可供vSphere管理员使用,并且已成为该产品版本6.0和6.5的痛点,因为它限制了基于业务需求或IT资源的整合而无需构建新vCenter Server的情况下进行合并或拆分的能力。组织。
VAMI界面已得到改进,新增了功能和选项卡,主要用于监视和故障排除:
VMware在此6.7版本的vSphere中简化了升级过程。从6.5升级到6.7将是第一个看到这种好处的人。VMware消除了主要版本升级通常需要的两次重新引导之一,因此您的升级将更快,过程更流畅,谁会抱怨呢?
这是一个很好的功能,至少在我的家庭实验室中,我还没有机会使用很多功能。对于初学者来说,您的硬件必须支持此功能,但是其想法是无需通过物理硬件重新启动过程即可重新启动虚拟机管理程序。目前仅支持有限数量的服务器,但将来数量会增加,因此我们至少期待更多。此功能不仅适用于重新引导,还适用于升级。系统将创建第二个ESXi内存映像,并且在重新引导主机时,它不会执行完全重新引导,而是切换到最新的。这是一个非常酷的新功能!
03 May 2018 by Ryan Birk
29
It’s been awhile since we’ve had a new vSphere release to get excited about but the wait is over – vSphere 6.7 has been officially released! I have been participating in the 6.7 beta so it’s finally nice to be able to talk about the release publicly. I wanted to put together a post that outlines some of the new features that are updated or new in this release. So let’s get to it!
Let’s talk first about what you can and can’t upgrade from. The new version supports upgrades and migrations from vSphere 6.0 or 6.5 only. If your environment is running version 5.5, the upgrade to 6.7 needs an additional step. You will have to migrate at least to version 6.0 then to 6.7. If you have a mixed environment with vCenter Server running 6.0 or 6.5 managing ESXi hosts 5.5, you have to upgrade the hosts to at least version 6.0 before upgrading the vCenter Server. VMware has given plenty of time and warnings about this and many saw it coming, but now it’s official, get off of 5.5 as soon as possible!
Currently supported migration paths to version 6.7:
VMware has once again updated some maximums. Most of the usual CPU and memory stuff stayed the same, but other changes are listed below. Fault Tolerant VMs get another upgrade and can be now bumped to 8 vCPUs, which is nice.
Virtual Machine Maximums | 6.5 | 6.7 |
Persistent Memory – NVDIMM controllers per VM | N/A | 1 |
Persistent Memory – Non-volatile memory per virtual machine | N/A | 1024GB |
Storage Virtual Adapters and Devices – Virtual SCSI targets per virtual SCSI adapter | 15 | 64 |
Storage Virtual Adapters and Devices – Virtual SCSI targets per virtual machine | 60 | 256 |
Networking Virtual Devices – Virtual RDMA Adapters per Virtual Machine | N/A | 1 |
ESXi Host Maximums | 6.5 | 6.7 |
Fault Tolerance maximums – Virtual CPUs per virtual machine | 4 | 8 |
Fault Tolerance maximums – RAM per FT VM | 64GB | 128GB |
Host CPU maximums – Logical CPUs per host | 576 | 768 |
ESXi Host Persistent Memory Maximums – Maximum Non-volatile memory per host | N/A | 1TB |
ESXi Host Memory Maximums – Maximum RAM per host | 12TB | 16TB |
Fibre Channel – Number of total paths on a server | 2048 | 4096 |
Common VMFS – Volumes per host | 512 | 1024 |
iSCSI Physical – LUNs per server | 512 | 1024 |
iSCSI Physical – Number of total paths on a server | 2048 | 4096 |
Fibre Channel – LUNs per host | 512 | 1024 |
Virtual Volumes – Number of PEs per host | 256 | 512 |
For me, this is hands down my favorite update in 6.7. I am constantly asked when they’re going to kill the Flash-based client and the answer is very soon with the announced deprecation of the Flash-based client last year. The HTML 5 client is now 95% feature complete. They’ve really polished the client and worked hard on it over the last several months. I’m very impressed with the work they’ve done here, so nice work to the vCenter team! The Platform Services Controller functionality is now part of the vSphere Client as well. It now located under the Administration menu, the PSC options are divided between two tabs. Certificate management has its own tab and all other management is under the configuration tab.
Some of the newer workflows in the updated vSphere Client release include:
Note that not all of the above workflows are 100% finished but VMware has been constantly updating the workflows with each update. It’s only a matter of time.
Built-in backup tools were built into 6.5, but easily being able to backup the appliance automatically was a challenge. As you can see below, you can now set up a basic backup schedule to backup your appliance configuration for a later restore. You can also set the number of backups to retain. I’d still recommend backing up your vCenter Linux Appliance with a third party backup tool like Altaro, just because you can never have enough backups!
It has returned! A new feature with vCenter 6.7 is the ability to re-point a vCenter Server to another Platform Services Controller node, that resides in an entirely different vSphere SSO domain. This functionality is insanely huge for domain consolidation, and also domain splitting if needed with the cmss-util tool.
This functionality was available to vSphere administrators in vSphere 5.5 and has been a pain point of versions 6.0 and 6.5 of the product as it constrained the ability to consolidate or split without building new vCenter Servers, based on business requirements or consolidation of IT resources within an organization.
New CMSSO-UTIL CLI utility allows repointing vCenter appliance across SSO domains.
The VAMI interface has been improved with new features and tabs mainly focused on monitoring and troubleshooting:
VMware has streamlined the upgrade process in this 6.7 release of vSphere. Upgrades from 6.5 to 6.7 will be the first to see this benefit. VMware has eliminated one of the two reboots that were normally required for major version upgrades, so your upgrades are going to be faster and a smoother process, and who could complain there?!
This is a neat feature that I have not had the opportunity to use much of yet, at least in my home lab. For starters, your hardware has to support this feature, but the idea is that the hypervisor can be restarted without going through the physical hardware reboot process. There is currently only a limited number of servers that are supported but the number will be increased in the future, so we get to look forward to more of this at least. This feature works not only for reboots but also for upgrades. The system creates a second ESXi memory image and when the host is rebooted it doesn’t perform a full reboot but it switches over to the latest instead. This is a pretty cool new feature!
Copyright © 2003-2013 www.wpsshop.cn 版权所有,并保留所有权利。